Lab 3-Clustering
1. There are 4 variables in the dataset. Plot the scatter plot of the data using the following
pairs of variables (using Species as Group role):
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
2. Setting the number of clusters to 3, and the variables selected for clustering to (like in part
(d) of instructions:
For each case, produce the scatter plot, with _SEGMENT_ is assigned the role of Group
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
3. Repeat Question 2, but this time without data normalization. This means, for each
clustering task, set the Internal Standardization to None. Comment on why results of this
clustering analysis are different from those obtained in Question 2.
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
The results of this clustering analysis are different from those obtained in Question 2 because with
Standardization, the number of items in each segment is different. Also, not standardizing or
1
1. There are 4 variables in the dataset. Plot the scatter plot of the data using the following
pairs of variables (using Species as Group role):
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
2. Setting the number of clusters to 3, and the variables selected for clustering to (like in part
(d) of instructions:
For each case, produce the scatter plot, with _SEGMENT_ is assigned the role of Group
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
3. Repeat Question 2, but this time without data normalization. This means, for each
clustering task, set the Internal Standardization to None. Comment on why results of this
clustering analysis are different from those obtained in Question 2.
a. Sepal Width and Sepal Length b. Petal Width and Sepal Length c. Petal Width and Sepal Width
The results of this clustering analysis are different from those obtained in Question 2 because with
Standardization, the number of items in each segment is different. Also, not standardizing or
1
